There are a number of important similarities between the
selenodeiodinases as well as distinct differences (Figure 1 and
Table 1). First, as would be expected from the fact that these
are integral membrane proteins, all three contain a hydrophobic
sequence which is located in the amino-terminal region of all
three proteins. Topological studies of D1 have shown that this is
a single membrane spanning domain(14). The topology of the D2 and
D3 proteins has not yet been elucidated. The area of highest
similarity between the three enzymes surrounds the active center
which contains selenocysteine. Presumably these residues are
critical to the maintenance of the iodine transfer process and
Se- regeneration. In addition to these common features, the
carboxyterminal portions of the proteins resemble one another
much more closely than do their amino-termini.
Figure 1:
Figure 2:
Comparison of the two outer-ring deiodinases which activate T4 D1 catalyzes the removal of an iodine from either the inner or the outer-ring of T4 with approximately equal efficiency, but reverse T3 and T3-sulfate are the preferred substrates (Table 1). The Km of D2 for T4 is 1000-fold lower than is that of D1 and T4 is its preferred substrate. The proposed enzymatic schema for T4 deiodination by D1 is shown in Figure 2 (5,15,16). The reaction follows ping-pong kinetics with the iodothyronine and an as yet unidentified sulhydryl-containing co-factor, perhaps GSH, as the co- substrates. Deiodination by D1 is inhibited by PTU since this reagent competes with the sulfhydryl cofactor in the regeneration of the enzyme. D2 is only minimally sensitive to PTU which may be a function of the lower enzyme turnover number as well as structural reasons. The transcription rate of the D1 gene is increased by T3 and two thyroid hormone response elements have been identified in the 5' flanking region of the human D1 gene (17). D1 activity in liver and kidney is high in the hyperthyroid state which is important in understanding the increase in the effectiveness of PTU to inhibit T4 to T3 conversion in hyperthyroidism in humans. D2 mRNA levels, in contrast, are reduced in the hyperthyroid state and increased in hypothyroidism (18). D2 is the most recently cloned of the three deiodinases and we know least about its structural-functional features (18-20). One of the most surprising aspects of D2 are the differences in tissue specific expression between rats and humans. D2 mRNA is expressed in rat and human CNS, pituitary, brown fat, and placenta, but also in human, but not in rat, thyroid, skeletal, and cardiac muscle (18,19,21). The presence of D2 activity in human skeletal muscle in particular raises questions about the accuracy of our current concepts regarding peripheral thyroid hormone activation in humans which have been based largely on experiments in rats. Thus, while we had previously thought that D2 provides primarily intracellular T3, we now must consider the possibilty that in humans it can provide plasma T3 as well. Production of plasma T3 by D2 has previously been documented in the newborn rat and chronically cold-exposed adult rat presumably from D2 in brown fat (22). The presence of high levels of D2 in human thyroid is especially interesting and unexpected (21) D2 mRNA is markedly higher in Graves' than in normal thyroid tissues. In some Graves' thyroids, virtually all the T3 produced from T4 is generated by D2 rather than by the D1 which is also present in this tissue (21). What is the relative role of D1 and D2 in the production of plasma T3 in humans? The wider distribution of D2 in humans than in the rat, particularly its presence in skeletal muscle, has suggested that this low Km deiodinase may play a more important role in this process than it does in the rat. This concept is supported by a review of a number of earlier studies in the context of recently available data. The first is the issue of the sensitivity of peripheral T4 to T3 conversion in humans to the D1-specific inhibitor, PTU. Early studies showed that in the hyperthyroid patient with Graves' disease, PTU in large doses can inhibit about 50% of T4 to T3 conversion (23). On the other hand, in euthyroid subjects there is inhibition of less than 20% of this process (24,25). Since D1 activity is increased by T3, and it is a high Km enzyme, we can explain these apparently conflicting results by the concept that D1 catalyzed T4 to T3 conversion is a more important source of plasma T3 in the hyperthyroid state while the PTU-insensitive D2 is more important in the euthyroid state. Furthermore, the fraction of T4 converted to T3 in humans increases in hypothyroidism or when serum T4 concentrations are reduced (26,27). This would be the expected result if D2, an enzyme which is increased in the low T4 or hypothyroid state, can generate a significant quantity of plasma T3. D2 is a much more efficient enzyme for T4 deiodination than is D1 the activity of which is decreased in the hypothyroid state. Thus one might speculate that in situations characterized by normal or especially low serum T4 concentrations, with or without hypothyroidism, D2 would play an important role in the generation of plasma T3. Examples of these would be mild to moderate hypothyroidism, iodine deficiency, and hyperthyroid patients under treatment with thioamide drugs. On the other hand, in the untreated hyperthyroid patient, peripheral T3 production would be predominantly via D1 catalyzed T4 5' deiodination. What role could the recently recognized D2 activity in human thyroid tissue play in this process? While the answer to this question is not known there are certain situations in which it could play a significant role. The most likely scenario in which thyroidal D2 would contribute to plasma T3 would be in the patient with Graves' disease in whom antithyroid drug treatment, especially with PTU, has reduced circulating T4 to normal or even reduced levels. Such patients would have an enlarged thyroid expressing high levels of D2 and, in addition, the low levels of T4 which favor D2-catalyzed T3 production. One might speculate that the D2 in Graves' thyroid might even have access to extrathyroidal T4 although this is by no means established. If so, it would contribute to the high circulating T3 to T4 ratios typical of such patients. The patterns of T3 production in humans in different conditions are summarized in Table 2. Therein, I have speculated on the proportional contributions to plasma T3 of T3 from thyroglobulin (Tg), and that from D1 and D2 catalyzed T4 5' deiodination in hyperthyroid, euthyroid, and hypothyroid states as well as in iodine deficiency. The latter is the condition in which D2-catalyzed T3 production is likely to be most significant. Table 2: Likely Sources of Circulating T3 in Humans: A Reassessment
At this time, we have only begun to explore the processes regulating thyroid hormone activation and inactivation especially in humans. The next few years should result in a major increase of our knowledge consequent to the application of modern molecular biological techniques to the study of the physiological role of these interesting and unusual proteins. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
